Python guide

Install the client from the Datamonger checkout, starting at the repository root:

python -m pip install ./packages/python

Inspect dataset metadata before fetching data:

from datamonger import data_info, fetch_data

info = data_info("uci:iris@1")
print(info.license)
result = fetch_data("uci:iris@1", return_info=True)
print(result.info.dataset_id)
print(result.info.registry_release, result.info.registry_index_sha256)
print(result.info.canonical_digest)

Dataset operations also accept separate arguments, such as fetch_data("iris", source="uci", version="1"). Omit source for qualified references and omit version when the reference embeds it. Duplicate arguments raise an error even when the values agree. "uci:iris" selects the registry's declared default version.

The default registry index is bundled. Fetching an uncached dataset needs network access. Record an explicit dataset version and a strong registry selector for reproducibility, and check dataset license and provenance metadata.

Use the Python section in the navigation bar to browse the extracted API reference.

See the complete Python client guide for return types, registry selection, offline use, and cache management.

Return to the Datamonger overview.